Publications

Gelatinase A transcriptional regulation is the consequence of combinatorial interactions with key promoter and enhancer elements identified within this gene. A potent 40 bp enhancer response element, RE-1, located in the near 5' flanking regions of the rat and human gelatinase A genes drives high-level expression in glomerular mesangial cells (MCs). Southwestern-blot analysis of MC nuclear extracts revealed specific interactions of RE-1 with at least four proteins, of which three have been identified as p53, activator protein 2 and the single-stranded DNA-binding factor Y-box protein-1 (YB-1). In the present study, we report the identification of a fourth 17 kDa RE-1-binding protein as the rat homologue (nm23-beta) of the human nm23-H1 metastasis suppressor gene. Recombinant nm23-beta protein bound only the single-stranded forms of the RE-1 sequence. Mutagenesis revealed direct interaction of nm23-beta with a repeat sequence, 5'-GGGTTT-3', shown previously to specifically interact with YB-1 [Mertens, Harendza, Pollock and Lovett (1997) J. Biol. Chem. 272, 22905-22912], and recombinant nm23-beta protein competed for single-stranded YB-1 binding. Transient transfection of MC with an nm23-beta expression plasmid within the context of a RE-1/simian virus 40 promoter/luciferase reporter yielded a concentration-dependent repression (80-90%) of luciferase activity in MC and Rat1 fibroblasts. A similar pattern of nm23-beta repression was demonstrated within the context of the RE-1/homologous gelatinase A promoter. Co-transfection of nm23-beta blocked YB-1-mediated activation of transcription and expression of gelatinase A. Nm23-beta may be an important physiological regulator of gelatinase A transcription that acts by competitive interference with the single-stranded transactivator YB-1. Gelatinase A is a key mediator of tumour metastasis, suggesting that competitive suppression of transcription by nm23-beta (or the human nm23-H1) may be a component of the reduced metastatic capabilities of cells expressing high levels of this protein.

Steatosis has emerged as a histologic finding of importance to the progression of hepatitis C virus (HCV)-associated liver disease. However, most studies of HCV-associated steatosis have excluded alcohol drinkers and individuals with diabetes and thus have not addressed the relative contribution of known causes of steatosis to liver injury in HCV-associated disease. To address this issue, we studied 297 consecutive patients with HCV who met inclusion criteria. Alcohol consumption, demographics, and serologic tests were correlated with degrees of steatosis and fibrosis on liver biopsy. Liver biopsy specimens were also examined for evidence of significant alcohol or nonalcoholic steatohepatitis (NASH) injury. In univariate analysis, steatosis correlated with type 2 diabetes mellitus (P =.005) and body mass index (BMI) (P =.0001) but not with the intensity of alcohol intake (in grams per day). In multivariate analysis, BMI (P =.0002) and genotype 3a infection (P =.02) were independent predictors of steatosis. When patients with risk factors for NASH were excluded, genotype 3a infection was the only independent predictor of steatosis. Steatosis (P =.04) and inflammation (P <.0001) scores on liver biopsy were the only independent predictors of fibrosis. Significant alcohol or NASH injury was found in only 6% of biopsy specimens. In conclusion, steatosis in HCV infection is associated with risk factors for NASH, particularly obesity, rather than alcohol consumption.